We have a system that loads data and then conducts data QC in PostgreSQL. The QC function's performance fluctuates drastically in one of our environments with no apparent pattern. I was able to track down the performance of the following simple query in the QC function:
WITH foo AS (SELECT full_address, jsonb_agg (gad_rec_id) gad_rec_ids
FROM azgiv.v_full_addresses
WHERE gad_gly_id = 495
GROUP BY full_address
HAVING count(1) > 1)
SELECT gad_nguid, gad_rec_id, foo.full_address
FROM azgiv.v_full_addresses JOIN foo
ON foo.full_address = v_full_addresses.full_address
AND v_full_addresses.gad_gly_id = 495;
When I ran into slow-performance situation (Fig 2), I had to ANALYZE the table behind the view before the query plan changes to fast (Fig 1). The v_full_addresses is a simple view of a partitioned table with bunch of columns concatenated.
Here are two images of the query plans for the above query. I am newbie when comes to understanding query optimization and any help is greatly appreciated.
&
If performance improves after you ANALYZE a table, that means that the database's knowledge about the distribution of the data is outdated.
The best remedy is to tell PostgreSQL to collect these statistics more often:
ALTER TABLE some_table SET (autovacuum_analyze_scale_factor = 0.02);
0.02 is five times lower than the default 0.1, so statistics will be gathered five times more often.
If the bad query plans are generated right after a bulk load, you must choose a different strategy. In this case the problem is that it takes up to a minute for auto-analyze to kick in and calculate new statistics.
In that case you should run an explicit ANALYZE at the end of the bulk load.
Related
I have this query that takes a very long time on my database. This SQL is generated from an ORM (Hibernate) inside of an application. I don't have access to the source code.
I was wondering if anyone can take a look at the following ANALYZE EXPLAIN output and suggest any Postgres tweaks I can make.
I don't know where to start or how to tune my database to service this query.
The query looks like this
select
resourceta0_.RES_ID as col_0_0_
from
HFJ_RESOURCE resourceta0_
left outer join HFJ_RES_LINK myresource1_ on resourceta0_.RES_ID = myresource1_.TARGET_RESOURCE_ID
left outer join HFJ_SPIDX_DATE myparamsda2_ on resourceta0_.RES_ID = myparamsda2_.RES_ID
left outer join HFJ_SPIDX_TOKEN myparamsto3_ on resourceta0_.RES_ID = myparamsto3_.RES_ID
where
(myresource1_.SRC_RESOURCE_ID in ('4954427' ... many more))
and myparamsda2_.HASH_IDENTITY=`5247847184787287691` and
(myparamsda2_.SP_VALUE_LOW>='1950-07-01 11:30:00' or myparamsda2_.SP_VALUE_HIGH>='1950-07-01 11:30:00')
and myparamsda2_.HASH_IDENTITY='5247847184787287691'
and (myparamsda2_.SP_VALUE_LOW<='1960-06-30 12:29:59.999' or myparamsda2_.SP_VALUE_HIGH<='1960-06-30 12:29:59.999')
and (myparamsto3_.HASH_VALUE in ('-5305902187566578701'))
limit '500'
And the execution plan looks like this: https://explain.depesz.com/s/EJgOq
Edit - updated to add the depesz link.
Edit 2 - added more information about the query.
The cause for the slowness are the bad row count estimates which make PostgreSQL choose a nested loop join. Almost all your time is spent in the index scan on hfj_res_link, which is repeated 1113 times.
My first attempt would be to ANALYZE hfj_spidx_date and see if that helps. If yes, make sure that autoanalyze treats that table more frequently.
The next attempt would be to
SET default_statistics_target = 1000;
and then ANALYZE as above. If that helps, use ALTER TABLE to increase the STATISTICS on the hash_identity and sp_value_high columns.
If that doesn't help either, and you have a recent version of PostgreSQL, you could try extended statistics:
CREATE STATISTICS myparamsda2_stats (dependencies)
ON hash_identity, sp_value_high FROM hfj_spidx_date;
Then ANALYZE the table again and see if that helps.
If all that doesn't help, and you cannot get the estimates correct, you have to try a different angle:
CREATE INDEX ON hfj_res_link (target_resource_id, src_resource_id);
That should speed up the index scan considerably and give you good response times.
Finally, if none of the above has any effect, you could use the cruse measure of disallowing nested loop joins for this query:
BEGIN;
SET LOCAL enable_nestloop = off;
SELECT /* your query goes here */;
COMMIT;
I'm in the process of benchmarking some queries in redshift so that I can say something intelligent about changes I've made to a table, such as adding encodings and running a vacuum. I can query the stl_query table with a LIKE clause to find the queries I'm interested in, so I have the query id, but tables/views like stv_query_summary are much too granular and I'm not sure how to generate the summarization I need!
The gui dashboard shows the metrics I'm interested in, but the format is difficult to store for later analysis/comparison (in other words, I want to avoid taking screenshots). Is there a good way to rebuild that view with sql selects?
To add to Alex answer, I want to comment that stl_query table has the inconvenience that if the query was in a queue before the runtime then the queue time will be included in the run time and therefore the runtime won't be a very good indicator of performance for the query.
To understand the actual runtime of the query, check on stl_wlm_query for the total_exec_time.
select total_exec_time
from stl_wlm_query
where query='query_id'
There are some usefuls tools/scripts in https://github.com/awslabs/amazon-redshift-utils
Here is one of said scripts stripped out to give you query run times in milliseconds. Play with the filters, ordering etc to show the results you are looking for:
select userid, label, stl_query.query, trim(database) as database, trim(querytxt) as qrytext, starttime, endtime, datediff(milliseconds, starttime,endtime)::numeric(12,2) as run_milliseconds,
aborted, decode(alrt.event,'Very selective query filter','Filter','Scanned a large number of deleted rows','Deleted','Nested Loop Join in the query plan','Nested Loop','Distributed a large number of rows across the network','Distributed','Broadcasted a large number of rows across the network','Broadcast','Missing query planner statistics','Stats',alrt.event) as event
from stl_query
left outer join ( select query, trim(split_part(event,':',1)) as event from STL_ALERT_EVENT_LOG group by query, trim(split_part(event,':',1)) ) as alrt on alrt.query = stl_query.query
where userid <> 1
-- and (querytxt like 'SELECT%' or querytxt like 'select%' )
-- and database = ''
order by starttime desc
limit 100
I currently have tables that are partitioned out by year & month for our sales transactions. For example, we have sales tables that would look something like this:
factdailysales_201501
factdailysales_201502
factdailysales_201503 etc ...
Generally, I've always performed dynamic SQL to capture a Start Date, End Date, find out what partitions those are, and then loop through each of those partitions ... but its starting to become such a hassle and I've learned that this is probably not the best way to do it in terms of just maintenance, trouble shooting, and performance.
I decided to build a view that would UNION ALL of my sales partitions together. However, I don't want selecting from the view to have to scan all of the partitions on execution, it would take away the whole purpose of partitioning tables out. Because of this, I added check constraints on date to each of my sales tables. This way when I selected from the view, it would know which tables to access from instead of scanning every table.
Here are the following examples below:
SELECT SUM([retail])
FROM Sales_Orig
WHERE [Date] >= '2015-03-01'
This query has the execution plan of only pulling from the partitions that I need.
My problem that i'm facing right now is that most of the time when my team will be writing stored procedures, they would more than likely write their queries where a date variable is passed into the where statement.
DECLARE #SD DATE = '2015-03-01'
SELECT SUM([retail])
FROM Sales_Orig
WHERE [Date] >= #SD
However, when a variable is being passed in, the execution plan now scans ALL of the partitions in the view, causing the performance to take wayyy longer than when I hard coded in the date
I suppose I could do dynamic SQL again and insert the date string into the SELECT statement, but it would bring me back to the beginning of trying to get rid of dynamic SQL in the first place for this simple sales query.
So my question is, am I setting this up wrong? Am I on the right track? It seems that the view can't take in a variable for the check constraint and ends up scanning every table. Is there another approach anyone would recommend? Maybe my original solution of just looping through partitions via dynamic SQL is the best way to do it?
** EDIT **
http://sqlsunday.com/2014/08/31/partitioned-views/
This article is actually where I initially saw the idea! It seems when using that exact same solution, I'm still experiencing the same struggle!
Thanks!!
Okay this might work. It's a table-valued function that only access tables according to your #start and #end parameters so only accessing your "partitions" that it needs. I figured you could take this concept and write some dynamic SQL to create all the if statements.
Now of course new tables are added every day so how does that tie in. Well I think the best way would be is that every day you alter the function adding the next sales table. That way querying it is simple. And you could use the same dynamic sql you used to create the function to alter it which should be relatively simple.
Note: I added default values that are the min and max of the data type DATE. That way you could query something like everything from 20140101 and onward or vice versa.
Your tables
SELECT CAST('20150101' AS DATE) datesVal INTO factDailySales_20150101;
SELECT CAST('20150102' AS DATE) datesVal INTO factDailySales_20150102;
SELECT CAST('20150103' AS DATE) datesVal INTO factDailySales_20150103;
The Function
CREATE FUNCTION ufn_factTotalSales (#Start DATE = '17530101', #End DATE = '99991231')
RETURNS #factTotalSales TABLE
(
datesVal DATE
)
AS
BEGIN
IF(CAST('20150101' AS DATE) BETWEEN #Start AND #End)
BEGIN
INSERT INTO #factTotalSales
SELECT datesVal
FROM factDailySales_20150101
END
IF(CAST('20150102' AS DATE) BETWEEN #Start AND #End)
BEGIN
INSERT INTO #factTotalSales
SELECT datesVal
FROM factDailySales_20150102
END
IF(CAST('20150103' AS DATE) BETWEEN #Start AND #End)
BEGIN
INSERT INTO #factTotalSales
SELECT datesVal
FROM factDailySales_20150103
END
RETURN;
END
GO
All tables
SELECT *
FROM ufn_factTotalSales(default,default)
All tables greater than or equal to 20150102
SELECT *
FROM ufn_factTotalSales('20150102',default)
**All tables less than or equal to 20150102
SELECT *
FROM ufn_factTotalSales(default,'20150102')
All tables between specific range
SELECT *
FROM ufn_factTotalSales('20150101','20150102')
Is this the ideal solution? No. The ideal would be to combine all tables into one and having good indexes. I know you said that wouldn't work because of the way other code has been written. Hear me out. Now perhaps this is off the wall, lets say you do combine the tables but obviously there are old scripts looking for specific daily sales tables. Maybe you could create views with the dailySales names that access the factTotalSales. OR You could create synonyms for the factTotalSales that would correspond to each factDailySales.
Maybe you could look into that. It wouldn't be easy, but I think letting SQL Server optimize your queries the way it was designed is a better way of doing it instead of forcing it with dynamic SQL.
Just my two cents. Hope this helps. At the very least, I hope it gave you some ideas.
5 years later: option(recompile).
The planner needs to have access to the constants to eliminate the table entirely from the query plan. With a variable, without a forced recompile, a generic plan is used. (Related: parameter sniffing.)
While this means the query plan is larger as it has to include all tables, it does not mean that all tables are actually scanned: look at the IO stats, as table scan elimination occurs even if such shows in the query plan.
The 'Number Of Executions' in the query plan will be 0 when the tables are not scanned: unfortunately, these branches are still reported as a non-zero percentage cost "Table Scan" node in the query plan & UI, which will appear high proportionally if the query is trivially fast. The displayed percentage cost of these extra "Table Scan" nodes approaches zero as the amount of data returned from the actually used base tables increases.
This same optimization/elimination occurs when the view is not a Partitioned View (eg. base tables are missing partition column in PK), yet the underlying tables have a suitable Check Constraint on the filtered column. It also occurs when the view selects a constant value to establish the partition that is not otherwise stored in the table. With a constant in the query or recompiled plan the tables will be eliminated entirely. With a variable the tables will still not actually be scanned and thus eliminated logically during query execution.
The use of a proper Partitioned View is only really beneficial to allow a direct Insert & Update, with the major caveat that it requires the partition column to be in each table's PK and disallows the use of an identity column (making a Partitioned View largely useless IMOHO). SQL Server handles the optimizations very similarly for other quasi-Partitioned View cases.
(This is on SQL Server 2014; earlier versions might not have optimized the different patterns as efficiently.)
To start off, I want to do an estimated count of how large a table is. Since I am building an analytical data of the database using graphs, exact count is not important. Thus I came across this wiki and it suggested doing an estimated count using
SELECT reltuples FROM pg_class WHERE relname = 'table_name';
Now in order to get an updated count, we would have to do an analyze on that table.
So my question is that is using analyze to get an updated count of the reltuples the same thing as doing count(*)? Is there a performance hit on doing analyze as much as doing count(*)?
The cost of running analyze depends on how much of the table is sampled, which depends on the default_statistics_target or the per-column statistics setting. It could be either faster or slower than count(*) depending on your specifics--there is not replacement for actually trying and seeing on your own system with our own data.
But normally if you are happy with an estimate, you would just use the reltuples you find there already. re-analyzing the table each time would seem to defeat the purpose.
Am I safe to assume that where I have stored procedures using the tempdb to write a temporary table, I'd be better off switching these to table variables to get better performance?
Temp tables are better in performance. If you use a Table Variable and the Data in the Variable gets too big, the SQL Server converts the Variable automatically into a temp table.
It depends, like almost every Database related question, on what you try to do. So it is hard to answer without more information.
So my answer is, try it and have a look at the execution plan. Use the fastest way with the lowest costs.
MSDN - Displaying Graphical Execution Plans (SQL Server Management Studio)
#Table can be faster as there is less "setup time" since the object is in memory only.
#Tables have a lot of catches though.
You can have a primary key on a #Table but thats about it. Other indexes Clustered NonClustered for combinations of columns are not possible.
Also if your table is going to contain any real data volumes (more then about 200 maybe 1000 rows) then accessing the table will be slower. Especially when you will probably not have a useful index on it.
#Tables are a pain in procs as they need to be dropped when debugging, They take longer to create. and they take longer to setup as you need to add indexs as a second step. But if you have lots of data then its #tables every time.
Even in cases where you have less then 100 rows of data in a table you may still want to use #Tables as you can create a usefull index on the table.
In summary i use #Tables most of the time for the ease when doing simple proc etc. But anything that need to perform should be a #Table.
#Tables have no statistics so the execution plan entails more guesswork. Hence the recommended upper limit of 1000-ish rows. #Tables have statistics but these can be cached between invocations. If your cardinalities differ significantly each time the SP runs you'd want to REBUILD and RECOMPILE each time. This is an overhead, of course, but one which must be balanced against the cost of a rubbish plan.
Both types will do IO to TempDB.
So no, #Tables are not a panacea.
Table variables can perform very poorly as the number of rows in them increases.
Why is this?
Table variables don’t have distribution statistics and don’t trigger recompiles. Because of this, SQL Server is not able to estimate the number of rows in a table variable like it does for normal tables. When the optimiser compiles code that contains a table variable, it assumes a table is empty and uses an expected row count of 1 for the cardinality estimate. Because the optimiser only thinks a table variable contains a single row, it picks operators for the execution plan that work well with a small set of records, like the NESTED LOOPS operator for a JOIN operation.
As an example, I have just fixed a stored procedure which was performing poorly. The code was populating a table variable and using it in a join to filter the number of rows to accounts which were relevant:
FROM dbo.DimInvestorAccount
INNER JOIN #accounts acclist
ON acclist.AccountNumber = DimInvestorAccount.investorAccountNumber
+ 9 additional tables joined...
When run for list of 1700 accounts, the query was taking 1m17s. Just changing the filter table definition from:
DECLARE #accounts TABLE (AccountNumber VARCHAR(20) COLLATE Latin1_General_BIN INDEX idx NONCLUSTERED)
to
CREATE TABLE #accounts (AccountNumber VARCHAR(20) COLLATE Latin1_General_BIN INDEX idx NONCLUSTERED)
brought the query time down to 800ms. Note that with 5 rows in the table, there was no significant difference - both temp table and table variable run in +/-400ms.
Microsoft's recommendation is to use Table Variables if the number of rows is <100.
Note that Microsoft have made changes in SQL Server 2019 to improve this (v15.x/Compatibility level 150)